Two 2022 timing forecasts for V393 Car have RMS errors of 168 and 1678 seconds in ten later TESS sectors

We test two central pulse timing curves published for V393 Car in 2022. The curves combine a circular orbit with either a linear or a quadratic pulse ephemeris. We measure 20 pulse phases from ten later public TESS sectors in 2023 and 2025. We use a fixed main pulse frequency and fit three harmonics. Only original data set the correction to the phase zero and clock rate. Our original-data extractions match the published observed timings to 1.87 s root mean square (RMS) for V393 Car and 1.30 s RMS for Chang 134. In the later V393 Car data, the two fixed curves have RMS errors of 168.43 s and 1678.14 s. The median block jackknife measurement error is 3.52 s. Changes in flux processing, harmonic count, and interval length give similar forecast errors. Synthetic tests check phase sign, delay recovery, and bias from the real sampling windows. A limited period sensitivity test also leaves large errors. The evidence shows that these two fixed curves do not predict the later phases. We do not calculate a full posterior prediction interval, determine the physical cause, exclude other companion models, or claim a new planet. The package includes exact light-curve columns, source hashes, 20 new timing measurements, figures, code, and instructions for an offline rerun.

V393 Car · TIC 364399376 · TESS · pulsation timing · phase modulation · forecast test · delta Scuti · substellar companions · reproducible astronomy · Chang 134